Location: RemoteFluent Language Skills Required: English & Portuguese (global, excluding Brazilian Portuguese). Native fluency in English and Portuguese (global, excluding Brazilian Portuguese) is required for this position.Why This Role ExistsAt Mercor, we believe the safest AI is the one that's already been attacked — by us. We are assembling a red team for this project - human data experts who probe AI models with adversarial inputs, surface vulnerabilities, and generate the red team data that makes AI safer for our customers.This project involves reviewing AI outputs that touch on sensitive topics such as bias, misinformation, or harmful behaviors. All work is text-based, and participation in higher-sensitivity projects is optional and supported by clear guidelines and wellness resources. Before being exposed to any content, the topics will be clearly communicated.What You'll DoRed team conversational AI models and agents: jailbreaks, prompt injections, misuse cases, bias exploitation, multi-turn manipulationGenerate high-quality human data: annotate failures, classify vulnerabilities, and flag systemic risksApply structure: follow taxonomies, benchmarks, and playbooks to keep testing consistentDocument reproducibly: produce reports, datasets,



and attack cases customers can act onWho You AreYou bring prior red teaming experience (AI adversarial work, cybersecurity, socio-technical probing)You're curious and adversarial: you instinctively push systems to breaking pointsYou're structured: you use frameworks or benchmarks, not just random hacksYou're communicative: you explain risks clearly to technical and non-technical stakeholdersYou're adaptable: thrive on moving across projects and customersNice-to-Have SpecialtiesAdversarial ML: jailbreak datasets, prompt injection, RLHF/DPO attacks, model extractionCybersecurity: penetration testing, exploit development, reverse engineeringSocio-technical risk: harassment/disinfo probing, abuse analysis, conversational AI testingCreative probing: psychology, acting, writing for unconventional adversarial thinkingWhat Success Looks LikeYou uncover vulnerabilities automated tests missYou deliver reproducible artifacts that strengthen customer AI systemsEvaluation coverage expands: more scenarios tested, fewer surprises in productionMercor customers trust the safety of their AI because you've already probed it like an adversaryWhy Join MercorBuild experience in human data-driven AI red teaming at the frontier of safetyPlay a direct role in making AI systems more robust, protected, and trustworthy

📌 Ai Safety Specialist - Fully Remote (Edmonton)
🏢 Mercor
📍 Edmonton